Logo ROOT   6.18/05
Reference Guide
VariablePCATransform.cxx
Go to the documentation of this file.
1// @(#)root/tmva $Id$
2// Author: Andreas Hoecker, Joerg Stelzer, Helge Voss, Eckhard von Toerne
3
4/**********************************************************************************
5 * Project: TMVA - a Root-integrated toolkit for multivariate data analysis *
6 * Package: TMVA *
7 * Class : VariablePCATransform *
8 * Web : http://tmva.sourceforge.net *
9 * *
10 * Description: *
11 * Implementation (see header for description) *
12 * *
13 * Authors (alphabetical): *
14 * Andreas Hoecker <Andreas.Hocker@cern.ch> - CERN, Switzerland *
15 * Peter Speckmayer <Peter.Speckmayer@cern.ch> - CERN, Switzerland *
16 * Joerg Stelzer <Joerg.Stelzer@cern.ch> - CERN, Switzerland *
17 * Eckhard v. Toerne <evt@uni-bonn.de> - U of Bonn, Germany *
18 * Helge Voss <Helge.Voss@cern.ch> - MPI-K Heidelberg, Germany *
19 * *
20 * Copyright (c) 2005-2011: *
21 * CERN, Switzerland *
22 * MPI-K Heidelberg, Germany *
23 * U. of Bonn, Germany *
24 * *
25 * Redistribution and use in source and binary forms, with or without *
26 * modification, are permitted according to the terms listed in LICENSE *
27 * (http://tmva.sourceforge.net/LICENSE) *
28 **********************************************************************************/
29
30/*! \class TMVA::VariablePCATransform
31\ingroup TMVA
32Linear interpolation class
33*/
34
36
37#include "TMVA/DataSet.h"
38#include "TMVA/Event.h"
39#include "TMVA/MsgLogger.h"
40#include "TMVA/Tools.h"
41#include "TMVA/Types.h"
42
43#include "TMatrixD.h"
44#include "TMatrixDBase.h"
45#include "TPrincipal.h"
46#include "TVectorD.h"
47#include "TVectorF.h"
48
49#include <iostream>
50#include <iomanip>
51#include <stdexcept>
52#include <algorithm>
53
55
56////////////////////////////////////////////////////////////////////////////////
57/// constructor
58
60: VariableTransformBase( dsi, Types::kPCA, "PCA" )
61{
62}
63
64////////////////////////////////////////////////////////////////////////////////
65/// destructor
66
68{
69 for (UInt_t i=0; i<fMeanValues.size(); i++) {
70 if (fMeanValues.at(i) != 0) delete fMeanValues.at(i);
71 if (fEigenVectors.at(i) != 0) delete fEigenVectors.at(i);
72 }
73}
74
75////////////////////////////////////////////////////////////////////////////////
76/// initialization of the transformation.
77/// Has to be called in the preparation and not in the constructor,
78/// since the number of classes it not known at construction, but
79/// only after the creation of the DataSet which might be later.
80
82{
83}
84
85////////////////////////////////////////////////////////////////////////////////
86/// calculate the principal components using the ROOT class TPrincipal
87/// and the normalization
88
90{
91 Initialize();
92
93 if (!IsEnabled() || IsCreated()) return kTRUE;
94
95 Log() << kINFO << "Preparing the Principle Component (PCA) transformation..." << Endl;
96
97 UInt_t inputSize = fGet.size();
98
99 SetNVariables(inputSize);
100
101 // TPrincipal doesn't support PCA transformation for 1 or less variables
102 if (inputSize <= 1) {
103 Log() << kFATAL << "Cannot perform PCA transformation for " << inputSize << " variable only" << Endl;
104 return kFALSE;
105 }
106
107 if (inputSize > 200) {
108 Log() << kINFO << "----------------------------------------------------------------------------"
109 << Endl;
110 Log() << kINFO
111 << ": More than 200 variables, will not calculate PCA!" << Endl;
112 Log() << kINFO << "----------------------------------------------------------------------------"
113 << Endl;
114 return kFALSE;
115 }
116
117 CalculatePrincipalComponents( events );
118
119 SetCreated( kTRUE );
120
121 return kTRUE;
122}
123
124////////////////////////////////////////////////////////////////////////////////
125/// apply the principal component analysis
126
128{
129 if (!IsCreated()) return 0;
130
131 // const Int_t inputSize = fGet.size();
132 // const UInt_t nCls = GetNClasses();
133
134 // if we have more than one class, take the last PCA analysis where all classes are combined if
135 // the cls parameter is outside the defined classes
136 // If there is only one class, then no extra class for all events of all classes has to be created
137
138 //if (cls < 0 || cls > GetNClasses()) cls = (fMeanValues.size()==1?0:2);//( GetNClasses() == 1 ? 0 : 1 ); ;
139 // EVT this is a workaround to address the reader problem with transforma and EvaluateMVA(std::vector<float/double> ,...)
140 if (cls < 0 || cls >= (int) fMeanValues.size()) cls = fMeanValues.size()-1;
141 // EVT workaround end
142
143 // Perform PCA and put it into PCAed events tree
144
145 if (fTransformedEvent==0 ) {
146 fTransformedEvent = new Event();
147 }
148
149 std::vector<Float_t> input;
150 std::vector<Char_t> mask;
151 std::vector<Float_t> principalComponents;
152
153 Bool_t hasMaskedEntries = GetInput( ev, input, mask );
154
155 if( hasMaskedEntries ){ // targets might be masked (for events where the targets have not been computed yet)
156 UInt_t numMasked = std::count(mask.begin(), mask.end(), (Char_t)kTRUE);
157 UInt_t numOK = std::count(mask.begin(), mask.end(), (Char_t)kFALSE);
158 if( numMasked>0 && numOK>0 ){
159 Log() << kFATAL << "You mixed variables and targets in the decorrelation transformation. This is not possible." << Endl;
160 }
161 SetOutput( fTransformedEvent, input, mask, ev );
162 return fTransformedEvent;
163 }
164
165 X2P( principalComponents, input, cls );
166 SetOutput( fTransformedEvent, principalComponents, mask, ev );
167
168 return fTransformedEvent;
169}
170
171////////////////////////////////////////////////////////////////////////////////
172/// apply the principal component analysis
173/// TODO: implementation of inverse transformation
174/// Log() << kFATAL << "Inverse transformation for PCA transformation not yet implemented. Hence, this transformation cannot be applied together with regression. Please contact the authors if necessary." << Endl;
175
177{
178 if (!IsCreated()) return 0;
179 // const Int_t inputSize = fGet.size();
180 const UInt_t nCls = GetNClasses();
181 //UInt_t evCls = ev->GetClass();
182
183 // if we have more than one class, take the last PCA analysis where all classes are combined if
184 // the cls parameter is outside the defined classes
185 // If there is only one class, then no extra class for all events of all classes has to be created
186 if (cls < 0 || UInt_t(cls) > nCls) cls = (fMeanValues.size()==1?0:2);//( GetNClasses() == 1 ? 0 : 1 ); ;
187 // Perform PCA and put it into PCAed events tree
188
189 if (fBackTransformedEvent==0 ) fBackTransformedEvent = new Event();
190
191 std::vector<Float_t> principalComponents;
192 std::vector<Char_t> mask;
193 std::vector<Float_t> output;
194
195 GetInput( ev, principalComponents, mask, kTRUE );
196 P2X( output, principalComponents, cls );
197 SetOutput( fBackTransformedEvent, output, mask, ev, kTRUE );
198
199 return fBackTransformedEvent;
200}
201
202////////////////////////////////////////////////////////////////////////////////
203/// calculate the principal components for the signal and the background data
204/// it uses the MakePrincipal method of ROOT's TPrincipal class
205
206void TMVA::VariablePCATransform::CalculatePrincipalComponents( const std::vector< Event*>& events )
207{
208 UInt_t nvars = 0, ntgts = 0, nspcts = 0;
209 CountVariableTypes( nvars, ntgts, nspcts );
210 if( nvars>0 && ntgts>0 )
211 Log() << kFATAL << "Variables and targets cannot be mixed in PCA transformation." << Endl;
212
213 const Int_t inputSize = fGet.size();
214
215 // if we have more than one class, add another PCA analysis which combines all classes
216 const UInt_t nCls = GetNClasses();
217 const UInt_t maxPCA = (nCls<=1) ? nCls : nCls+1;
218
219 // PCA [signal/background/class x/class y/... /all classes]
220 std::vector<TPrincipal*> pca(maxPCA);
221 for (UInt_t i=0; i<maxPCA; i++) pca[i] = new TPrincipal(nvars,"");
222
223 // !! Not normalizing and not storing input data, for performance reasons. Should perhaps restore normalization.
224 // But this can be done afterwards by adding a normalisation transformation (user defined)
225
226 Long64_t ievt, entries = events.size();
227 Double_t *dvec = new Double_t[inputSize];
228
229 std::vector<Float_t> input;
230 std::vector<Char_t> mask;
231 for (ievt=0; ievt<entries; ievt++) {
232 const Event* ev = events[ievt];
233 UInt_t cls = ev->GetClass();
234
235 Bool_t hasMaskedEntries = GetInput( ev, input, mask );
236 if (hasMaskedEntries){
237 Log() << kWARNING << "Print event which triggers an error" << Endl;
238 std::ostringstream oss;
239 ev->Print(oss);
240 Log() << oss.str();
241 Log() << kFATAL << "Masked entries found in event read in when calculating the principal components for the PCA transformation." << Endl;
242 }
243
244 UInt_t iinp = 0;
245 for( std::vector<Float_t>::iterator itInp = input.begin(), itInpEnd = input.end(); itInp != itInpEnd; ++itInp )
246 {
247 Float_t value = (*itInp);
248 dvec[iinp] = (Double_t)value;
249 ++iinp;
250 }
251
252 pca.at(cls)->AddRow( dvec );
253 if (nCls > 1) pca.at(maxPCA-1)->AddRow( dvec );
254 }
255
256 // delete possible leftovers
257 for (UInt_t i=0; i<fMeanValues.size(); i++) if (fMeanValues[i] != 0) delete fMeanValues[i];
258 for (UInt_t i=0; i<fEigenVectors.size(); i++) if (fEigenVectors[i] != 0) delete fEigenVectors[i];
259 fMeanValues.resize(maxPCA,0);
260 fEigenVectors.resize(maxPCA,0);
261
262 for (UInt_t i=0; i<maxPCA; i++ ) {
263 pca.at(i)->MakePrincipals();
264
265 // retrieve mean values, eigenvectors and sigmas
266 fMeanValues[i] = new TVectorD( *(pca.at(i)->GetMeanValues()) ); // need to copy since we want to own
267 fEigenVectors[i] = new TMatrixD( *(pca.at(i)->GetEigenVectors()) );
268 }
269
270 for (UInt_t i=0; i<maxPCA; i++) delete pca.at(i);
271 delete [] dvec;
272}
273
274////////////////////////////////////////////////////////////////////////////////
275/// Calculate the principal components from the original data vector
276/// x, and return it in p (function extracted from TPrincipal::X2P)
277/// It's the users responsibility to make sure that both x and p are
278/// of the right size (i.e., memory must be allocated for p)
279
280void TMVA::VariablePCATransform::X2P( std::vector<Float_t>& pc, const std::vector<Float_t>& x, Int_t cls ) const
281{
282 const Int_t nInput = x.size();
283 pc.assign(nInput,0);
284
285 for (Int_t i = 0; i < nInput; i++) {
286 Double_t pv = 0;
287 for (Int_t j = 0; j < nInput; j++)
288 pv += (((Double_t)x.at(j)) - (*fMeanValues.at(cls))(j)) * (*fEigenVectors.at(cls))(j,i);
289 pc[i] = pv;
290 }
291}
292
293////////////////////////////////////////////////////////////////////////////////
294/// Perform the back-transformation from the principal components
295/// pc, and return x
296/// It's the users responsibility to make sure that both x and pc are
297/// of the right size (i.e., memory must be allocated for p)
298
299void TMVA::VariablePCATransform::P2X( std::vector<Float_t>& x, const std::vector<Float_t>& pc, Int_t cls ) const
300{
301 const Int_t nInput = pc.size();
302 x.assign(nInput,0);
303
304 for (Int_t i = 0; i < nInput; i++) {
305 Double_t xv = 0;
306 for (Int_t j = 0; j < nInput; j++)
307 xv += (((Double_t)pc.at(j)) * (*fEigenVectors.at(cls))(i,j) ) + (*fMeanValues.at(cls))(j);
308 x[i] = xv;
309 }
310}
311
312////////////////////////////////////////////////////////////////////////////////
313/// write mean values to stream
314
316{
317 for (Int_t sbType=0; sbType<2; sbType++) {
318 o << "# PCA mean values " << std::endl;
319 const TVectorD* means = fMeanValues[sbType];
320 o << (sbType==0 ? "Signal" : "Background") << " " << means->GetNrows() << std::endl;
321 for (Int_t row = 0; row<means->GetNrows(); row++) {
322 o << std::setprecision(12) << std::setw(20) << (*means)[row];
323 }
324 o << std::endl;
325 }
326 o << "##" << std::endl;
327
328 // write eigenvectors to stream
329 for (Int_t sbType=0; sbType<2; sbType++) {
330 o << "# PCA eigenvectors " << std::endl;
331 const TMatrixD* mat = fEigenVectors[sbType];
332 o << (sbType==0 ? "Signal" : "Background") << " " << mat->GetNrows() << " x " << mat->GetNcols() << std::endl;
333 for (Int_t row = 0; row<mat->GetNrows(); row++) {
334 for (Int_t col = 0; col<mat->GetNcols(); col++) {
335 o << std::setprecision(12) << std::setw(20) << (*mat)[row][col] << " ";
336 }
337 o << std::endl;
338 }
339 }
340 o << "##" << std::endl;
341}
342
343////////////////////////////////////////////////////////////////////////////////
344/// create XML description of PCA transformation
345
347 void* trfxml = gTools().AddChild(parent, "Transform");
348 gTools().AddAttr(trfxml, "Name", "PCA");
349
351
352 // write mean values to stream
353 for (UInt_t sbType=0; sbType<fMeanValues.size(); sbType++) {
354 void* meanxml = gTools().AddChild( trfxml, "Statistics");
355 const TVectorD* means = fMeanValues[sbType];
356 gTools().AddAttr( meanxml, "Class", (sbType==0 ? "Signal" :(sbType==1 ? "Background":"Combined")) );
357 gTools().AddAttr( meanxml, "ClassIndex", sbType );
358 gTools().AddAttr( meanxml, "NRows", means->GetNrows() );
359 TString meansdef = "";
360 for (Int_t row = 0; row<means->GetNrows(); row++)
361 meansdef += gTools().StringFromDouble((*means)[row]) + " ";
362 gTools().AddRawLine( meanxml, meansdef );
363 }
364
365 // write eigenvectors to stream
366 for (UInt_t sbType=0; sbType<fEigenVectors.size(); sbType++) {
367 void* evxml = gTools().AddChild( trfxml, "Eigenvectors");
368 const TMatrixD* mat = fEigenVectors[sbType];
369 gTools().AddAttr( evxml, "Class", (sbType==0 ? "Signal" :(sbType==1 ? "Background":"Combined") ) );
370 gTools().AddAttr( evxml, "ClassIndex", sbType );
371 gTools().AddAttr( evxml, "NRows", mat->GetNrows() );
372 gTools().AddAttr( evxml, "NCols", mat->GetNcols() );
373 TString evdef = "";
374 for (Int_t row = 0; row<mat->GetNrows(); row++)
375 for (Int_t col = 0; col<mat->GetNcols(); col++)
376 evdef += gTools().StringFromDouble((*mat)[row][col]) + " ";
377 gTools().AddRawLine( evxml, evdef );
378 }
379}
380
381////////////////////////////////////////////////////////////////////////////////
382/// Read the transformation matrices from the xml node
383
385{
386 Int_t nrows, ncols;
387 UInt_t clsIdx;
388 TString classtype;
389 TString nodeName;
390
391 Bool_t newFormat = kFALSE;
392
393 void* inpnode = NULL;
394
395 inpnode = gTools().GetChild(trfnode, "Selection"); // new xml format
396 if( inpnode!=NULL )
397 newFormat = kTRUE; // new xml format
398
399 if( newFormat ){
400 // ------------- new format --------------------
401 // read input
403
404 }
405
406 void* ch = gTools().GetChild(trfnode);
407 while (ch) {
408 nodeName = gTools().GetName(ch);
409 if (nodeName == "Statistics") {
410 // read mean values
411 gTools().ReadAttr(ch, "Class", classtype);
412 gTools().ReadAttr(ch, "ClassIndex", clsIdx);
413 gTools().ReadAttr(ch, "NRows", nrows);
414
415 // set the correct size
416 if (fMeanValues.size()<=clsIdx) fMeanValues.resize(clsIdx+1,0);
417 if (fMeanValues[clsIdx]==0) fMeanValues[clsIdx] = new TVectorD( nrows );
418 fMeanValues[clsIdx]->ResizeTo( nrows );
419
420 // now read vector entries
421 std::stringstream s(gTools().GetContent(ch));
422 for (Int_t row = 0; row<nrows; row++) s >> (*fMeanValues[clsIdx])(row);
423 }
424 else if ( nodeName == "Eigenvectors" ) {
425 // Read eigenvectors
426 gTools().ReadAttr(ch, "Class", classtype);
427 gTools().ReadAttr(ch, "ClassIndex", clsIdx);
428 gTools().ReadAttr(ch, "NRows", nrows);
429 gTools().ReadAttr(ch, "NCols", ncols);
430
431 if (fEigenVectors.size()<=clsIdx) fEigenVectors.resize(clsIdx+1,0);
432 if (fEigenVectors[clsIdx]==0) fEigenVectors[clsIdx] = new TMatrixD( nrows, ncols );
433 fEigenVectors[clsIdx]->ResizeTo( nrows, ncols );
434
435 // now read matrix entries
436 std::stringstream s(gTools().GetContent(ch));
437 for (Int_t row = 0; row<nrows; row++)
438 for (Int_t col = 0; col<ncols; col++)
439 s >> (*fEigenVectors[clsIdx])[row][col];
440 } // done reading eigenvectors
441 ch = gTools().GetNextChild(ch);
442 }
443
444 SetCreated();
445}
446
447////////////////////////////////////////////////////////////////////////////////
448/// Read mean values from input stream
449
450void TMVA::VariablePCATransform::ReadTransformationFromStream( std::istream& istr, const TString& classname )
451{
452 char buf[512];
453 istr.getline(buf,512);
454 TString strvar, dummy;
455 Int_t nrows(0), ncols(0);
456 UInt_t classIdx=(classname=="signal"?0:1);
457
458 for (UInt_t i=0; i<fMeanValues.size(); i++) {
459 if (fMeanValues.at(i) != 0) delete fMeanValues.at(i);
460 if (fEigenVectors.at(i) != 0) delete fEigenVectors.at(i);
461 }
462 fMeanValues.resize(3);
463 fEigenVectors.resize(3);
464
465 Log() << kINFO << "VariablePCATransform::ReadTransformationFromStream(): " << Endl;
466
467 while (!(buf[0]=='#'&& buf[1]=='#')) { // if line starts with ## return
468 char* p = buf;
469 while (*p==' ' || *p=='\t') p++; // 'remove' leading whitespace
470 if (*p=='#' || *p=='\0') {
471 istr.getline(buf,512);
472 continue; // if comment or empty line, read the next line
473 }
474 std::stringstream sstr(buf);
475 sstr >> strvar;
476 if (strvar=="signal" || strvar=="background") {
477
478 sstr >> nrows;
479 Int_t sbType = (strvar=="signal" ? 0 : 1);
480
481 if (fMeanValues[sbType] == 0) fMeanValues[sbType] = new TVectorD( nrows );
482 else fMeanValues[sbType]->ResizeTo( nrows );
483
484 // now read vector entries
485 for (Int_t row = 0; row<nrows; row++) istr >> (*fMeanValues[sbType])(row);
486
487 } // done reading vector
488
489 istr.getline(buf,512); // reading the next line
490 }
491
492 // Read eigenvectors from input stream
493 istr.getline(buf,512);
494 while (!(buf[0]=='#'&& buf[1]=='#')) { // if line starts with ## return
495 char* p = buf;
496 while(*p==' ' || *p=='\t') p++; // 'remove' leading whitespace
497 if (*p=='#' || *p=='\0') {
498 istr.getline(buf,512);
499 continue; // if comment or empty line, read the next line
500 }
501 std::stringstream sstr(buf);
502 sstr >> strvar;
503 if (strvar=="signal" || strvar=="background") {
504
505 // coverity[tainted_data_argument]
506 sstr >> nrows >> dummy >> ncols;
507 Int_t sbType = (strvar=="signal" ? 0 : 1);
508
509 if (fEigenVectors[sbType] == 0) fEigenVectors[sbType] = new TMatrixD( nrows, ncols );
510 else fEigenVectors[sbType]->ResizeTo( nrows, ncols );
511
512 // now read matrix entries
513 for (Int_t row = 0; row<fEigenVectors[sbType]->GetNrows(); row++) {
514 for (Int_t col = 0; col<fEigenVectors[sbType]->GetNcols(); col++) {
515 istr >> (*fEigenVectors[sbType])[row][col];
516 }
517 }
518
519 } // done reading matrix
520 istr.getline(buf,512); // reading the next line
521 }
522 fMeanValues[2] = new TVectorD( *fMeanValues[classIdx] );
523 fEigenVectors[2] = new TMatrixD( *fEigenVectors[classIdx] );
524
525 SetCreated();
526}
527
528////////////////////////////////////////////////////////////////////////////////
529/// creates C++ code fragment of the PCA transform for inclusion in standalone C++ class
530
531void TMVA::VariablePCATransform::MakeFunction( std::ostream& fout, const TString& fcncName,
532 Int_t part, UInt_t trCounter, Int_t )
533{
534 UInt_t nvar = fEigenVectors[0]->GetNrows();
535
536 // creates a PCA transformation function
537 UInt_t numC = fMeanValues.size();
538 if (part==1) {
539 fout << std::endl;
540 fout << " void X2P_"<<trCounter<<"( const double*, double*, int ) const;" << std::endl;
541 fout << " double fMeanValues_"<<trCounter<<"["<<numC<<"]["
542 << fMeanValues[0]->GetNrows() << "];" << std::endl; // mean values
543 fout << " double fEigenVectors_"<<trCounter<<"["<<numC<<"]["
544 << fEigenVectors[0]->GetNrows() << "]["
545 << fEigenVectors[0]->GetNcols() <<"];" << std::endl; // eigenvectors
546 fout << std::endl;
547 }
548
549 // sanity check
550 if (numC>1){
551 if (fMeanValues[0]->GetNrows() != fMeanValues[1]->GetNrows() ||
552 fEigenVectors[0]->GetNrows() != fEigenVectors[1]->GetNrows() ||
553 fEigenVectors[0]->GetNcols() != fEigenVectors[1]->GetNcols()) {
554 Log() << kFATAL << "<MakeFunction> Mismatch in vector/matrix dimensions" << Endl;
555 }
556 }
557
558 if (part==2) {
559
560 fout << std::endl;
561 fout << "//_______________________________________________________________________" << std::endl;
562 fout << "inline void " << fcncName << "::X2P_"<<trCounter<<"( const double* x, double* p, int index ) const" << std::endl;
563 fout << "{" << std::endl;
564 fout << " // Calculate the principal components from the original data vector" << std::endl;
565 fout << " // x, and return it in p (function extracted from TPrincipal::X2P)" << std::endl;
566 fout << " // It's the users responsibility to make sure that both x and p are" << std::endl;
567 fout << " // of the right size (i.e., memory must be allocated for p)." << std::endl;
568 fout << " const int nVar = " << nvar << ";" << std::endl;
569 fout << std::endl;
570 fout << " for (int i = 0; i < nVar; i++) {" << std::endl;
571 fout << " p[i] = 0;" << std::endl;
572 fout << " for (int j = 0; j < nVar; j++) p[i] += (x[j] - fMeanValues_"<<trCounter<<"[index][j]) * fEigenVectors_"<<trCounter<<"[index][j][i];" << std::endl;
573 fout << " }" << std::endl;
574 fout << "}" << std::endl;
575 fout << std::endl;
576 fout << "//_______________________________________________________________________" << std::endl;
577 fout << "inline void " << fcncName << "::InitTransform_"<<trCounter<<"()" << std::endl;
578 fout << "{" << std::endl;
579 fout << " // PCA transformation, initialisation" << std::endl;
580
581 // fill vector of mean values
582 fout << " // initialise vector of mean values" << std::endl;
583 std::streamsize dp = fout.precision();
584 for (UInt_t index=0; index<numC; index++) {
585 for (int i=0; i<fMeanValues[index]->GetNrows(); i++) {
586 fout << " fMeanValues_"<<trCounter<<"["<<index<<"]["<<i<<"] = " << std::setprecision(12)
587 << (*fMeanValues[index])(i) << ";" << std::endl;
588 }
589 }
590
591 // fill matrix of eigenvectors
592 fout << std::endl;
593 fout << " // initialise matrix of eigenvectors" << std::endl;
594 for (UInt_t index=0; index<numC; index++) {
595 for (int i=0; i<fEigenVectors[index]->GetNrows(); i++) {
596 for (int j=0; j<fEigenVectors[index]->GetNcols(); j++) {
597 fout << " fEigenVectors_"<<trCounter<<"["<<index<<"]["<<i<<"]["<<j<<"] = " << std::setprecision(12)
598 << (*fEigenVectors[index])(i,j) << ";" << std::endl;
599 }
600 }
601 }
602 fout << std::setprecision(dp);
603 fout << "}" << std::endl;
604 fout << std::endl;
605 fout << "//_______________________________________________________________________" << std::endl;
606 fout << "inline void " << fcncName << "::Transform_"<<trCounter<<"( std::vector<double>& iv, int cls ) const" << std::endl;
607 fout << "{" << std::endl;
608 fout << " // PCA transformation" << std::endl;
609 fout << " const int nVar = " << nvar << ";" << std::endl;
610 fout << " double *dv = new double[nVar];" << std::endl;
611 fout << " double *rv = new double[nVar];" << std::endl;
612 fout << " if (cls < 0 || cls > "<<GetNClasses()<<") {"<< std::endl;
613 fout << " if ("<<GetNClasses()<<" > 1 ) cls = "<<GetNClasses()<<";"<< std::endl;
614 fout << " else cls = "<<(numC==1?0:2)<<";"<< std::endl;
615 fout << " }"<< std::endl;
616
617 VariableTransformBase::MakeFunction(fout, fcncName, 0, trCounter, 0 );
618
619 fout << " for (int ivar=0; ivar<nVar; ivar++) dv[ivar] = iv[indicesGet.at(ivar)];" << std::endl;
620
621 fout << std::endl;
622 fout << " // Perform PCA and put it into PCAed events tree" << std::endl;
623 fout << " this->X2P_"<<trCounter<<"( dv, rv, cls );" << std::endl;
624 fout << " for (int ivar=0; ivar<nVar; ivar++) iv[indicesPut.at(ivar)] = rv[ivar];" << std::endl;
625
626 fout << std::endl;
627 fout << " delete [] dv;" << std::endl;
628 fout << " delete [] rv;" << std::endl;
629 fout << "}" << std::endl;
630 }
631}
static RooMathCoreReg dummy
int Int_t
Definition: RtypesCore.h:41
char Char_t
Definition: RtypesCore.h:29
unsigned int UInt_t
Definition: RtypesCore.h:42
const Bool_t kFALSE
Definition: RtypesCore.h:88
bool Bool_t
Definition: RtypesCore.h:59
double Double_t
Definition: RtypesCore.h:55
long long Long64_t
Definition: RtypesCore.h:69
float Float_t
Definition: RtypesCore.h:53
const Bool_t kTRUE
Definition: RtypesCore.h:87
#define ClassImp(name)
Definition: Rtypes.h:365
TMatrixT< Double_t > TMatrixD
Definition: TMatrixDfwd.h:22
TVectorT< Double_t > TVectorD
Definition: TVectorDfwd.h:22
Class that contains all the data information.
Definition: DataSetInfo.h:60
UInt_t GetClass() const
Definition: Event.h:87
void Print(std::ostream &o) const
print method
Definition: Event.cxx:352
void * GetNextChild(void *prevchild, const char *childname=0)
XML helpers.
Definition: Tools.cxx:1174
void * AddChild(void *parent, const char *childname, const char *content=0, bool isRootNode=false)
add child node
Definition: Tools.cxx:1136
Bool_t AddRawLine(void *node, const char *raw)
XML helpers.
Definition: Tools.cxx:1202
TString StringFromDouble(Double_t d)
string tools
Definition: Tools.cxx:1245
void * GetChild(void *parent, const char *childname=0)
get child node
Definition: Tools.cxx:1162
const char * GetName(void *node)
XML helpers.
Definition: Tools.cxx:1194
void ReadAttr(void *node, const char *, T &value)
read attribute from xml
Definition: Tools.h:337
void AddAttr(void *node, const char *, const T &value, Int_t precision=16)
add attribute to xml
Definition: Tools.h:355
Singleton class for Global types used by TMVA.
Definition: Types.h:73
Linear interpolation class.
void CalculatePrincipalComponents(const std::vector< Event * > &)
calculate the principal components for the signal and the background data it uses the MakePrincipal m...
void P2X(std::vector< Float_t > &, const std::vector< Float_t > &, Int_t cls) const
Perform the back-transformation from the principal components pc, and return x It's the users respons...
virtual const Event * Transform(const Event *const, Int_t cls) const
apply the principal component analysis
virtual const Event * InverseTransform(const Event *const, Int_t cls) const
apply the principal component analysis TODO: implementation of inverse transformation Log() << kFATAL...
virtual ~VariablePCATransform(void)
destructor
Bool_t PrepareTransformation(const std::vector< Event * > &)
calculate the principal components using the ROOT class TPrincipal and the normalization
virtual void MakeFunction(std::ostream &fout, const TString &fncName, Int_t part, UInt_t trCounter, Int_t cls)
creates C++ code fragment of the PCA transform for inclusion in standalone C++ class
void Initialize()
initialization of the transformation.
void ReadTransformationFromStream(std::istream &, const TString &)
Read mean values from input stream.
void X2P(std::vector< Float_t > &, const std::vector< Float_t > &, Int_t cls) const
Calculate the principal components from the original data vector x, and return it in p (function extr...
VariablePCATransform(DataSetInfo &dsi)
constructor
void WriteTransformationToStream(std::ostream &) const
write mean values to stream
virtual void AttachXMLTo(void *parent)
create XML description of PCA transformation
virtual void ReadFromXML(void *trfnode)
Read the transformation matrices from the xml node.
Linear interpolation class.
virtual void MakeFunction(std::ostream &fout, const TString &fncName, Int_t part, UInt_t trCounter, Int_t cls)=0
getinput and setoutput equivalent
virtual void ReadFromXML(void *trfnode)=0
Read the input variables from the XML node.
virtual void AttachXMLTo(void *parent)=0
create XML description the transformation (write out info of selected variables)
Int_t GetNrows() const
Definition: TMatrixTBase.h:124
Int_t GetNcols() const
Definition: TMatrixTBase.h:127
Principal Components Analysis (PCA)
Definition: TPrincipal.h:20
Basic string class.
Definition: TString.h:131
Int_t GetNrows() const
Definition: TVectorT.h:75
Double_t x[n]
Definition: legend1.C:17
static constexpr double s
static constexpr double pc
void Initialize(Bool_t useTMVAStyle=kTRUE)
Definition: tmvaglob.cxx:176
Tools & gTools()
MsgLogger & Endl(MsgLogger &ml)
Definition: MsgLogger.h:158
Double_t Log(Double_t x)
Definition: TMath.h:748
static void output(int code)
Definition: gifencode.c:226